How do intimate relationships function in times of stress? To what extent can a partner make a difference in reducing stress-related anxiety and depression, and to what extent does the supportive role have an impact on the partner's own mental health? Are the answers to these questions the same for both men and women as supporters, and for relationships of different closeness and satisfaction? Although researchers have been studying questions such as these for decades, definitive answers are still wanting. One reason for the slow rate of progress is that methods for modeling stress and coping dynamics have been limited. The proposed five-year research program attempts to address these issues on two fronts, one methodological and one substantive. The methodological aim is to compare the strengths and weaknesses of recently developed daily diary designs to those of traditional panel survey designs with regard to measurement quality, sample selection and retention, and model specification biases. Laboratory experiments of measurement effects, computer simulation studies, and empirical comparisons of diary and panel study results will be used to address the aim. The substantive aim is to follow up leads from previous research that support is most effective when invisible to the stressed partner, and that contagion of the negative affect can be limited by the non-stressed partner's understanding of the causes of the affect. In addition, we propose to test the generalizability of our findings across gender of supporter, and across variation of relationship quality. These aims will be addressed using both daily diary and panel design data collected from 700 couples that contain one partner who is preparing for a major professional certification examination. The substantive findings will be critically examined in the context of the methodological results, and the methodological investigations will be informed by empirical facts from the field studies.